[PATCH v4 0/2] Add regulator support to qcom-ufs driver

[PATCH v4 0/2] Add regulator support to qcom-ufs driver

[Federico Amedeo Izzo via B4 Relay]

This series enables UFS support on SC7280 devices with UFS3.1.
SC7280 devices with UFS2.2 were already working.

The root cause was ABL leaving vcc-supply regulator disabled on UFS3.1
devices, and enabled on UFS2.2 devices that worked fine without
regulator support.

Signed-off-by: Federico Amedeo Izzo <federico@izzo.pro>
---
Changes in v4:
- Improve regulator mode fix: apply current mode during enable
- Remove regulator mode code from qcom-ufs driver
- Link to v3: https://patch.msgid.link/20260523-qcom-ufs-regulator-support-v3-0-1be2a80a2577@izzo.pro

Changes in v3:
- Fix bug causing HPM mode not being applied. Removed LPM/HPM hack
- Read regulator mode from dts property instead of hardcoding it
- Link to v2: https://patch.msgid.link/20260521-qcom-ufs-regulator-support-v2-1-d3ec82ae5dbc@izzo.pro

Changes in v2:
- Add vccq2-supply for UFS2.2 devices without internal eLDO regulator
- Fix wrong topic
- Link to v1: https://patch.msgid.link/20260521-dpu-add-dspp-gc-driver-v1-1-83372488cdcf@izzo.pro

To: Sumit Garg <sumit.garg@kernel.org>
To: u-boot-qcom@groups.io
To: u-boot@lists.denx.de
Cc: Jaehoon Chung <jh80.chung@samsung.com>
Cc: Peng Fan <peng.fan@nxp.com>
Cc: Casey Connolly <casey.connolly@linaro.org>
Cc: Neil Armstrong <neil.armstrong@linaro.org>
Cc: Tom Rini <trini@konsulko.com>
Cc: Luca Weiss <luca.weiss@fairphone.com>
Cc: Aswin Murugan <aswin.murugan@oss.qualcomm.com>
Cc: Federico Amedeo Izzo <federico@izzo.pro>
Cc: Bhupesh Sharma <bhupesh.linux@gmail.com>
Cc: Neha Malcom Francis <n-francis@ti.com>
Cc: Julien Stephan <jstephan@baylibre.com>
Cc: Marek Vasut <marek.vasut+renesas@mailbox.org>

---
Federico Amedeo Izzo (2):
      regulator: qcom-rpmh-regulator: fix regulator mode mismatch
      ufs: qcom: add ufs host regulator support

 drivers/power/regulator/qcom-rpmh-regulator.c | 106 ++++++++++++++------------
 drivers/ufs/ufs-qcom.c                        |  25 ++++++
 drivers/ufs/ufs-qcom.h                        |   4 +
 3 files changed, 85 insertions(+), 50 deletions(-)
---
base-commit: 744cf5d4e398c5f657e5c3eeb44d00fa5175dc85
change-id: 20260521-qcom-ufs-regulator-support-466ea6ad202a

Best regards,
--  
Federico Amedeo Izzo <federico@izzo.pro>

[PATCH v4 1/2] regulator: qcom-rpmh-regulator: fix regulator mode mismatch

[Federico Amedeo Izzo via B4 Relay]

From: Federico Amedeo Izzo <federico@izzo.pro>

Initial regulator mode was read from dts but never applied.
This caused a mismatch between saved mode and actual regulator mode.

Apply the current mode from priv->mode during enable() and move
rpmh_regulator_vrm_set_mode function before rpmh_regulator_set_enable_state().

Signed-off-by: Federico Amedeo Izzo <federico@izzo.pro>
---
 drivers/power/regulator/qcom-rpmh-regulator.c | 106 ++++++++++++++------------
 1 file changed, 56 insertions(+), 50 deletions(-)

diff --git a/drivers/power/regulator/qcom-rpmh-regulator.c b/drivers/power/regulator/qcom-rpmh-regulator.c
index 4d65aae1690..f789b5b6f86 100644
--- a/drivers/power/regulator/qcom-rpmh-regulator.c
+++ b/drivers/power/regulator/qcom-rpmh-regulator.c
@@ -295,6 +295,56 @@ static int rpmh_regulator_vrm_get_value(struct udevice *rdev)
 	return vreg->uv;
 }
 
+static int rpmh_regulator_vrm_set_mode_bypass(struct rpmh_vreg *vreg,
+					      unsigned int mode, bool bypassed)
+{
+	struct tcs_cmd cmd = {
+		.addr = vreg->addr + RPMH_REGULATOR_REG_VRM_MODE,
+	};
+	struct dm_regulator_mode *pmic_mode;
+	int i;
+
+	if (mode > REGULATOR_MODE_HPM)
+		return -EINVAL;
+
+	for (i = 0; i < vreg->hw_data->n_modes; i++) {
+		pmic_mode = &vreg->hw_data->pmic_mode_map[i];
+		if (pmic_mode->id == mode)
+			break;
+	}
+	if (pmic_mode->id != mode) {
+		printf("Invalid mode %d\n", mode);
+		return -EINVAL;
+	}
+
+	if (bypassed)
+		// XXX: should have a version check for PMIC4 but we don't have any yet
+		// and we don't use bypass mode
+		cmd.data = PMIC5_BOB_MODE_PASS;
+	else
+		cmd.data = pmic_mode->register_value;
+
+	return rpmh_regulator_send_request(vreg, &cmd, true);
+}
+
+static int rpmh_regulator_vrm_set_mode(struct udevice *rdev,
+				       int mode)
+{
+	struct rpmh_vreg *vreg = dev_get_priv(rdev);
+	int ret;
+
+	debug("%s: set_mode %d (current %d)\n", rdev->name, mode, vreg->mode);
+
+	if (mode == vreg->mode)
+		return 0;
+
+	ret = rpmh_regulator_vrm_set_mode_bypass(vreg, mode, vreg->bypassed);
+	if (!ret)
+		vreg->mode = mode;
+
+	return ret;
+}
+
 static int rpmh_regulator_is_enabled(struct udevice *rdev)
 {
 	struct rpmh_vreg *vreg = dev_get_priv(rdev);
@@ -331,6 +381,12 @@ static int rpmh_regulator_set_enable_state(struct udevice *rdev,
 	debug("%s: set_enable %d (current %d)\n", rdev->name, enable,
 	      vreg->enabled);
 
+	if (vreg->mode != -EINVAL) {
+		ret = rpmh_regulator_vrm_set_mode_bypass(vreg, vreg->mode, vreg->bypassed);
+		if (ret < 0)
+			return ret;
+	}
+
 	if (vreg->enabled == -EINVAL &&
 	    vreg->uv != -ENOTRECOVERABLE) {
 		ret = _rpmh_regulator_vrm_set_value(rdev,
@@ -346,56 +402,6 @@ static int rpmh_regulator_set_enable_state(struct udevice *rdev,
 	return ret;
 }
 
-static int rpmh_regulator_vrm_set_mode_bypass(struct rpmh_vreg *vreg,
-					      unsigned int mode, bool bypassed)
-{
-	struct tcs_cmd cmd = {
-		.addr = vreg->addr + RPMH_REGULATOR_REG_VRM_MODE,
-	};
-	struct dm_regulator_mode *pmic_mode;
-	int i;
-
-	if (mode > REGULATOR_MODE_HPM)
-		return -EINVAL;
-
-	for (i = 0; i < vreg->hw_data->n_modes; i++) {
-		pmic_mode = &vreg->hw_data->pmic_mode_map[i];
-		if (pmic_mode->id == mode)
-			break;
-	}
-	if (pmic_mode->id != mode) {
-		printf("Invalid mode %d\n", mode);
-		return -EINVAL;
-	}
-
-	if (bypassed)
-		// XXX: should have a version check for PMIC4 but we don't have any yet
-		// and we don't use bypass mode
-		cmd.data = PMIC5_BOB_MODE_PASS;
-	else
-		cmd.data = pmic_mode->register_value;
-
-	return rpmh_regulator_send_request(vreg, &cmd, true);
-}
-
-static int rpmh_regulator_vrm_set_mode(struct udevice *rdev,
-				       int mode)
-{
-	struct rpmh_vreg *vreg = dev_get_priv(rdev);
-	int ret;
-
-	debug("%s: set_mode %d (current %d)\n", rdev->name, mode, vreg->mode);
-
-	if (mode == vreg->mode)
-		return 0;
-
-	ret = rpmh_regulator_vrm_set_mode_bypass(vreg, mode, vreg->bypassed);
-	if (!ret)
-		vreg->mode = mode;
-
-	return ret;
-}
-
 static int rpmh_regulator_vrm_get_pmic_mode(struct rpmh_vreg *vreg, int *pmic_mode)
 {
 	struct tcs_cmd cmd = {

-- 
2.54.0

[PATCH v4 2/2] ufs: qcom: add ufs host regulator support

[Federico Amedeo Izzo via B4 Relay]

From: Federico Amedeo Izzo <federico@izzo.pro>

The qcom UFS driver works without touching the regulators on most platforms
as ABL leaves them already configured and enabled.
On some SC7280 phones with UFS 3.1 like nothing-spacewar, vcc regulator is
not enabled by ABL, so we need to configure and enable it to use the UFS
storage.

Tested on nothing-spacewar and motorola-dubai using Tauchgang U-Boot.

Signed-off-by: Federico Amedeo Izzo <federico@izzo.pro>
---
 drivers/ufs/ufs-qcom.c | 25 +++++++++++++++++++++++++
 drivers/ufs/ufs-qcom.h |  4 ++++
 2 files changed, 29 insertions(+)

diff --git a/drivers/ufs/ufs-qcom.c b/drivers/ufs/ufs-qcom.c
index dc40ee62daf..0e2e45f1075 100644
--- a/drivers/ufs/ufs-qcom.c
+++ b/drivers/ufs/ufs-qcom.c
@@ -16,6 +16,7 @@
 #include <generic-phy.h>
 #include <asm/gpio.h>
 #include <interconnect.h>
+#include <power/regulator.h>
 
 #include <linux/bitops.h>
 #include <linux/delay.h>
@@ -561,6 +562,19 @@ static int ufs_qcom_init(struct ufs_hba *hba)
 
 	priv->hba = hba;
 
+	/* enable regulators */
+	err = regulator_set_enable(priv->vcc, true);
+	if (err && err != -ENOSYS)
+		dev_warn(hba->dev, "failed to enable regulator vcc-supply:%d\n", err);
+
+	err = regulator_set_enable(priv->vccq, true);
+	if (err && err != -ENOSYS)
+		dev_warn(hba->dev, "failed to enable regulator vccq-supply:%d\n", err);
+
+	err = regulator_set_enable(priv->vccq2, true);
+	if (err && err != -ENOSYS)
+		dev_warn(hba->dev, "failed to enable regulator vccq2-supply:%d\n", err);
+
 	/* setup clocks */
 	ufs_qcom_setup_clocks(hba, true, PRE_CHANGE);
 
@@ -651,6 +665,17 @@ static int ufs_qcom_probe(struct udevice *dev)
 		dev_err(dev, "Warning: cannot get reset GPIO\n");
 	}
 
+	ret = device_get_supply_regulator(dev, "vcc-supply", &priv->vcc);
+	if (ret)
+		dev_warn(dev, "failed to get regulator vcc-supply:%d\n", ret);
+
+	ret = device_get_supply_regulator(dev, "vccq-supply", &priv->vccq);
+	if (ret)
+		dev_warn(dev, "failed to get regulator vccq-supply:%d\n", ret);
+
+	/* vccq2 is optional on UFS2 and missing on UFS3 and later */
+	device_get_supply_regulator(dev, "vccq2-supply", &priv->vccq2);
+
 	ret = ufshcd_probe(dev, &ufs_qcom_hba_ops);
 	if (ret) {
 		dev_err(dev, "ufshcd_probe() failed, ret:%d\n", ret);
diff --git a/drivers/ufs/ufs-qcom.h b/drivers/ufs/ufs-qcom.h
index de957ae60f3..115565db7e9 100644
--- a/drivers/ufs/ufs-qcom.h
+++ b/drivers/ufs/ufs-qcom.h
@@ -134,6 +134,10 @@ struct ufs_qcom_priv {
 	struct clk_bulk clks;
 	bool is_clks_enabled;
 
+	struct udevice *vcc;
+	struct udevice *vccq;
+	struct udevice *vccq2;
+
 	struct ufs_hw_version hw_ver;
 
 	/* Reset control of HCI */

-- 
2.54.0

Re: [PATCH v4 2/2] ufs: qcom: add ufs host regulator support

[Neha Malcom Francis]

On Sat, 23 May 2026 15:04:25 +0200, Federico Amedeo Izzo <federico@izzo.pro> wrote:
> diff --git a/drivers/ufs/ufs-qcom.c b/drivers/ufs/ufs-qcom.c
> index dc40ee62daf..0e2e45f1075 100644
> --- a/drivers/ufs/ufs-qcom.c
> +++ b/drivers/ufs/ufs-qcom.c
> @@ -561,6 +562,19 @@ static int ufs_qcom_init(struct ufs_hba *hba)
>  
>  	priv->hba = hba;
>  
> +	/* enable regulators */
> +	err = regulator_set_enable(priv->vcc, true);
> +	if (err && err != -ENOSYS)
> +		dev_warn(hba->dev, "failed to enable regulator vcc-supply:%d\n", err);
> +
> +	err = regulator_set_enable(priv->vccq, true);
> +	if (err && err != -ENOSYS)
> +		dev_warn(hba->dev, "failed to enable regulator vccq-supply:%d\n", err);
> +
> +	err = regulator_set_enable(priv->vccq2, true);
> +	if (err && err != -ENOSYS)
> +		dev_warn(hba->dev, "failed to enable regulator vccq2-supply:%d\n", err);

vccq2 is optional right? Is it verified that the optional code route will never
return ENOSYS and there is handling in regulator_set_enable for NULL regulators?

-- 
Neha Malcom Francis <n-francis@ti.com>

Re: [PATCH v4 2/2] ufs: qcom: add ufs host regulator support

[Federico Amedeo Izzo]

Il 2026-05-25 14:12 Neha Malcom Francis ha scritto:

> On Sat, 23 May 2026 15:04:25 +0200, Federico Amedeo Izzo 
> <federico@izzo.pro> wrote:
> 
>> diff --git a/drivers/ufs/ufs-qcom.c b/drivers/ufs/ufs-qcom.c
>> index dc40ee62daf..0e2e45f1075 100644
>> --- a/drivers/ufs/ufs-qcom.c
>> +++ b/drivers/ufs/ufs-qcom.c
>> @@ -561,6 +562,19 @@ static int ufs_qcom_init(struct ufs_hba *hba)
>> 
>> priv->hba = hba;
>> 
>> +    /* enable regulators */
>> +    err = regulator_set_enable(priv->vcc, true);
>> +    if (err && err != -ENOSYS)
>> +        dev_warn(hba->dev, "failed to enable regulator 
>> vcc-supply:%d\n", err);
>> +
>> +    err = regulator_set_enable(priv->vccq, true);
>> +    if (err && err != -ENOSYS)
>> +        dev_warn(hba->dev, "failed to enable regulator 
>> vccq-supply:%d\n", err);
>> +
>> +    err = regulator_set_enable(priv->vccq2, true);
>> +    if (err && err != -ENOSYS)
>> +        dev_warn(hba->dev, "failed to enable regulator 
>> vccq2-supply:%d\n", err);
> 
> vccq2 is optional right? Is it verified that the optional code route 
> will never
> return ENOSYS and there is handling in regulator_set_enable for NULL 
> regulators?

Hi,
vccq2 is indeed optional, if it's not present, regulator_set_enable() 
will handle the NULL regulator and return -ENOSYS.
The warning is skipped for -ENOSYS as we don't want to print warnings 
for non existing regulators,

Below is a section of regulator_set_enable() from regulator_uclass.c

int regulator_set_enable(struct udevice *dev, bool enable)
{
         const struct dm_regulator_ops *ops = dev_get_driver_ops(dev);
         struct dm_regulator_uclass_plat *uc_pdata;
         int ret, old_enable = 0;

         if (!ops || !ops->set_enable)
                 return -ENOSYS;
[...]

Greetings,
Federico Amedeo Izzo

Re: [PATCH v4 2/2] ufs: qcom: add ufs host regulator support

[Neha Malcom Francis]

On 2026-05-25 21:41:00+02:00, Federico Amedeo Izzo wrote:
> Il 2026-05-25 14:12 Neha Malcom Francis ha scritto:
> 
> > On Sat, 23 May 2026 15:04:25 +0200, Federico Amedeo Izzo 
> > <federico@izzo.pro> wrote:
> > 
> > 
> > vccq2 is optional right? Is it verified that the optional code route 
> > will never
> > return ENOSYS and there is handling in regulator_set_enable for NULL 
> > regulators?
> 
> Hi,
> vccq2 is indeed optional, if it's not present, regulator_set_enable() 
> will handle the NULL regulator and return -ENOSYS.
> The warning is skipped for -ENOSYS as we don't want to print warnings 
> for non existing regulators,
> 
> Below is a section of regulator_set_enable() from regulator_uclass.c
> 
> int regulator_set_enable(struct udevice *dev, bool enable)
> {
>          const struct dm_regulator_ops *ops = dev_get_driver_ops(dev);
>          struct dm_regulator_uclass_plat *uc_pdata;
>          int ret, old_enable = 0;
> 
>          if (!ops || !ops->set_enable)
>                  return -ENOSYS;
> [...]
> 

Cool that works. Thanks!

> Greetings,
> Federico Amedeo Izzo

Re: [PATCH v4 2/2] ufs: qcom: add ufs host regulator support

[Neha Malcom Francis]

On 2026-05-25 21:41:00+02:00, Federico Amedeo Izzo wrote:
> Il 2026-05-25 14:12 Neha Malcom Francis ha scritto:
> 
> > On Sat, 23 May 2026 15:04:25 +0200, Federico Amedeo Izzo 
> > <federico@izzo.pro> wrote:
> > 
> > 
> > vccq2 is optional right? Is it verified that the optional code route 
> > will never
> > return ENOSYS and there is handling in regulator_set_enable for NULL 
> > regulators?
> 
> Hi,
> vccq2 is indeed optional, if it's not present, regulator_set_enable() 
> will handle the NULL regulator and return -ENOSYS.
> The warning is skipped for -ENOSYS as we don't want to print warnings 
> for non existing regulators,
> 
> Below is a section of regulator_set_enable() from regulator_uclass.c
> 
> int regulator_set_enable(struct udevice *dev, bool enable)
> {
>          const struct dm_regulator_ops *ops = dev_get_driver_ops(dev);
>          struct dm_regulator_uclass_plat *uc_pdata;
>          int ret, old_enable = 0;
> 
>          if (!ops || !ops->set_enable)
>                  return -ENOSYS;
> [...]
>

Reviewed-by: Neha Malcom Francis <n-francis@ti.com>

> Greetings,
> Federico Amedeo Izzo

Re: [PATCH v4 0/2] Add regulator support to qcom-ufs driver

[Casey Connolly]

Hi Federico,

On 23/05/2026 15:04, Federico Amedeo Izzo via B4 Relay wrote:
> This series enables UFS support on SC7280 devices with UFS3.1.
> SC7280 devices with UFS2.2 were already working.
> 
> The root cause was ABL leaving vcc-supply regulator disabled on UFS3.1
> devices, and enabled on UFS2.2 devices that worked fine without
> regulator support.

Thanks a lot for working on this and finding the bug in the regulator
driver!

> 
> Signed-off-by: Federico Amedeo Izzo <federico@izzo.pro>

Reviewed-by: Casey Connolly <casey.connolly@linaro.org>

> ---
> Changes in v4:
> - Improve regulator mode fix: apply current mode during enable
> - Remove regulator mode code from qcom-ufs driver
> - Link to v3: https://patch.msgid.link/20260523-qcom-ufs-regulator-support-v3-0-1be2a80a2577@izzo.pro
> 
> Changes in v3:
> - Fix bug causing HPM mode not being applied. Removed LPM/HPM hack
> - Read regulator mode from dts property instead of hardcoding it
> - Link to v2: https://patch.msgid.link/20260521-qcom-ufs-regulator-support-v2-1-d3ec82ae5dbc@izzo.pro
> 
> Changes in v2:
> - Add vccq2-supply for UFS2.2 devices without internal eLDO regulator
> - Fix wrong topic
> - Link to v1: https://patch.msgid.link/20260521-dpu-add-dspp-gc-driver-v1-1-83372488cdcf@izzo.pro
> 
> To: Sumit Garg <sumit.garg@kernel.org>
> To: u-boot-qcom@groups.io
> To: u-boot@lists.denx.de
> Cc: Jaehoon Chung <jh80.chung@samsung.com>
> Cc: Peng Fan <peng.fan@nxp.com>
> Cc: Casey Connolly <casey.connolly@linaro.org>
> Cc: Neil Armstrong <neil.armstrong@linaro.org>
> Cc: Tom Rini <trini@konsulko.com>
> Cc: Luca Weiss <luca.weiss@fairphone.com>
> Cc: Aswin Murugan <aswin.murugan@oss.qualcomm.com>
> Cc: Federico Amedeo Izzo <federico@izzo.pro>
> Cc: Bhupesh Sharma <bhupesh.linux@gmail.com>
> Cc: Neha Malcom Francis <n-francis@ti.com>
> Cc: Julien Stephan <jstephan@baylibre.com>
> Cc: Marek Vasut <marek.vasut+renesas@mailbox.org>
> 
> ---
> Federico Amedeo Izzo (2):
>       regulator: qcom-rpmh-regulator: fix regulator mode mismatch
>       ufs: qcom: add ufs host regulator support
> 
>  drivers/power/regulator/qcom-rpmh-regulator.c | 106 ++++++++++++++------------
>  drivers/ufs/ufs-qcom.c                        |  25 ++++++
>  drivers/ufs/ufs-qcom.h                        |   4 +
>  3 files changed, 85 insertions(+), 50 deletions(-)
> ---
> base-commit: 744cf5d4e398c5f657e5c3eeb44d00fa5175dc85
> change-id: 20260521-qcom-ufs-regulator-support-466ea6ad202a
> 
> Best regards,
> --  
> Federico Amedeo Izzo <federico@izzo.pro>
> 
> 

-- 
// Casey (she/her)

Re: (subset) [PATCH v4 0/2] Add regulator support to qcom-ufs driver

[Casey Connolly]

On Sat, 23 May 2026 15:04:23 +0200, Federico Amedeo Izzo wrote:
> This series enables UFS support on SC7280 devices with UFS3.1.
> SC7280 devices with UFS2.2 were already working.
> 
> The root cause was ABL leaving vcc-supply regulator disabled on UFS3.1
> devices, and enabled on UFS2.2 devices that worked fine without
> regulator support.
> 
> [...]

Applied, thanks!

[1/2] regulator: qcom-rpmh-regulator: fix regulator mode mismatch
      https://source.denx.de/u-boot/custodians/u-boot-snapdragon/-/commit/5e6f370c1e3d

Best regards,
-- 
// Casey (she/they)

Re: [PATCH v4 0/2] Add regulator support to qcom-ufs driver

[Neil Armstrong]

Hi,

On Sat, 23 May 2026 15:04:23 +0200, Federico Amedeo Izzo wrote:
> This series enables UFS support on SC7280 devices with UFS3.1.
> SC7280 devices with UFS2.2 were already working.
> 
> The root cause was ABL leaving vcc-supply regulator disabled on UFS3.1
> devices, and enabled on UFS2.2 devices that worked fine without
> regulator support.
> 
> [...]

Thanks, Applied to https://source.denx.de/u-boot/custodians/u-boot-ufs (u-boot-ufs-next)

[1/2] regulator: qcom-rpmh-regulator: fix regulator mode mismatch
      (no commit info)
[2/2] ufs: qcom: add ufs host regulator support
      https://source.denx.de/u-boot/custodians/u-boot-ufs/-/commit/8591f01705b513ab51d6762bbabeaa0417033a95

-- 
Neil